the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Nov 2025
Large-scale drivers of compounding hot and dry events in three breadbasket regions
Abstract. Compound hot and dry events cause damage to ecosystems and society. While these events have been widely studied individually, their co-ocurrence and the associated large-scale atmospheric drivers remain less understood. Here, we use reanalysis products and observational data to identify compound hot and dry events in the historical period from 1960 to 2020. We analyze the large-scale circulation patterns associated with compound occurrence of hot and dry events when they affect large portions of three breadbasket regions in the Northern Hemisphere, namely North America, Europe and the Mediterranean and eastern Asia. We find that compound hot and dry events recur throughout the historical period and are consistently linked to Rossby wave patterns and mid-tropospheric anticyclonic ridging, which trigger land-atmosphere feedbacks resulting in the reinforcement of the events. Our study highlights that the spatial extent of compound hot and dry events offers a metric for assessing regional impacts.
- Preprint
(2318 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on egusphere-2025-5073', Yongli He, 08 Jan 2026
-
CC2: 'Reply on CC1', Natalia Castillo, 03 Feb 2026
1)Â Thanks for your comment. In Figure 5a, we are referring to the total number of COHDEs during high FA events in NA (not a high-low composite differences). The same applies to Figure 6a (EM) and 7a (AS). We will revise the caption of the figure accordingly to avoid ambiguity:
Figure 5. a) Total number of COHDES during high Fraction of Area (FA) events (19 events) in North America (NA, highlighted by the black contours). Composite differences of high FA events minus low FA events in NA: b) maximum daily temperature at 2m (shadings) and isohypses at 500 hPa (z500), c) precipitation (shadings) and isohypses at 500 hPa, d) surface sensible and e) surface latent heat fluxes, f) distribution of the dominant wavenumbers associated with the v250 anomalies, g) meridional component of the wind at 250 hPa (v250) and h) zonal component of the wind at 250 hPa (u250). Hatched areas in the composites indicate regions where differences are statistically significant at the 95% confidence level, based on a Student’s t-test.
Â
2) Thanks for your comment. This is due to the construction of the composite differences, which are based on events in North America only. Although Figure 5g shows a circumglobal wave train, the anomalies in the variables (precipitation, surface sensible and latent heat fluxes) triggering the compounding of hot and dry conditions are not strong enough in the places where the other ridges are located to trigger more compounding there.
Â
3) Thanks again for your comment. We were referring to Figure 1, as we describe the methodology in the flowchart and intended to indicate that this is the final step of the workflow. We will revise the manuscript to state this more clearly.
Â
Â
Citation: https://doi.org/10.5194/egusphere-2025-5073-CC2
-
CC2: 'Reply on CC1', Natalia Castillo, 03 Feb 2026
-
RC1: 'Comment on egusphere-2025-5073', Anonymous Referee #1, 19 Feb 2026
Castillo et al. investigate the occurrence of joint hot dry – events in three mid-latitude regions and provide a large-scale dynamical perspective by analysing Rossby wave composites. Compound events and the atmospheric drivers can be of high interest in the context of high impact low likelihood events such as concurrent breadbasket failures and therefore deserve some attention. The analysis is nicely done and the results are interesting. However, while the authors present a solid framework for investigating such events based on historical reanalysis the present manuscript falls short in delivering answers to some of the most pressing questions in this discourse, namely: To what extent is agricultural production affected (or at least regions of agricultural production) by hot dry events, is there a higher risk of concurrent breadbasket failures expected from atmosphere dynamical changes alone? To what degree do Rossby waves favour concurrent  hot-dry events across the mid-latitudes and is this effect important compared to other drivers/teleconnections, e.g. internal variability? What can we say about historical trends and future expectations? Answers to these questions lie at the fingertips of the authors, as most of the data seems to be provided (hot dry events, wave events, …) but a quantitative analysis is missing. I would therefore suggest major revisions to address my more specific comments below.
Major: l. 91 More recent drought studies usually rely on the SPEI instead of the SPI as it acknowledges the role of evapotranspiration due to high temperatures (e.g. Gebrechorkos et al. 2025), the effect that the authors name as the driving force connecting warming and more severe hot-dry events in l. 37 and the fact that authors want to investigate ‘physical drivers’. Results should not depend on this choice substantially, but it would be important to understand the differences using an example. E.g. is this why no trends are detected (Fig. 4), while the authors argue in their intro that combined hot-dry extremes become more relevant in a warmer climate? A section to motivate the indicator choice explicitly in the text would be useful as well.
l.147 it seems as if the selection of regions is not directly informed by local agricultural production as done e.g in Kornhuber et al. 2023 or Vogel et al. 2019. Thus, strictly speaking these are just ‘SREX regions’ and not necessarily ‘breadbasket regions’. The analysis is therefore fairly close to earlier studies from Sarhadi et al. 2019, Zhou et al. 2023 and Nasong et al. 2025, that are not cited in the present manuscript. It remains open how much of the global production of the named cereal crops is accounted for by focussing on these three specific areas, that include large ocean regions and areas that do not feature relevant agricultural activity (e.g. Northern Africa), but leave out Eastern Europe (while the Russian Heatwave is discussed as the prime example in the introduction) and India/Pakistan which was affected by a heatwave in 2022 as well (arguably in Spring not in Summer, see Aadhar & Mishra 2023) with severe agricultural impacts. Another open question would be, why were other mid-lat. regions in the northern hemisphere left out? The selection of these specific locations will almost certainly impose Rossby waves as a driving pattern as they are symmetrically distanced across the mid-latitudes. E.g. choosing zonally elongated regions will favour lower wavenumbers that relate to surface anomalies of a larger extent zonal extent, compared to higher wavenumbers.
Could the authors reflect more quantitatively on the relationship between hot/dry events and wave events? Is their concurrence significant? How do Wave events that relate to hot/dry events different from those that don’t relate to hot dry events? Could there be other driver(s) that preconditions the events?
One question that remains unanswered is the joint likelihood of seeing all three regions affected by hot/dry events and to what degree the large-scale circulation plays a role. Â
Minor:
l. 16 ‘Our study highlights that the spatial extent..’ as a last sentence for an abstract this could be more to the point. Do the authors refer to the hemispheric localisation or the size of the individual anomalies.
l. 37 a few references would be good here, as this would support the main claim of this sentence.
l.58 two papers that investigated Rossby Waves, hot-dry events and their impacts on Agriculture and Ecosystems seem to be missing here, possibly putting the authors claim that these have not been investigated in combination so far in perspective:
- Lian, X., Li, Y., Liu, J. et al. Northern ecosystem productivity reduced by Rossby-wave-driven hot–dry conditions. Nat. Geosci. 18, 615–623 (2025). https://doi.org/10.1038/s41561-025-01722-3
- Kornhuber, K., Lesk, C., Schleussner, C.F. et al. Risks of synchronized low yields are underestimated in climate and crop model projections. Nat Commun 14, 3528 (2023). https://doi.org/10.1038/s41467-023-38906-7
l. 58 the authors likely wanted to cite Raymond et al. 2022 here?
l.61 the research questions could be a bit clearer, what do the authors mean with ‘characteristics’, what do they mean with ‘large circulation patterns’ if not Rossby waves, in particular as the authors seem to limit their analysis to a Rossby wave metric?
Figure 2, if different colors and scattertypes are used, please add a legend or at least a description in the caption.
Figures: this might be a compression artefact when converting to .pdf but figures should be provided in higher resolution.
Figure 4. Can something be said about the trends? Why are statistics for JJA not shown? As these form the basis for Fig. 3 this would be good to see.
l. 235 this abbreviation has not been introduced yet?
Fig. 5 how are events aligned, drought / heat are captured on a 3 months basis, while v-winds/waves are detected daily?
Â
Â
Â
Citation: https://doi.org/10.5194/egusphere-2025-5073-RC1 -
RC2: 'Comment on egusphere-2025-5073', Anonymous Referee #2, 24 Mar 2026
This manuscript provides a nice overview of the linkage between compound hot and dry events and the atmospheric circulation. The authors showcase the well documented typical atmospheric patterns associated with such events - strong anticyclones centered of the region of hot and dry conditions, with the often expected strong surface heating through surface sensible heat flux. I enjoyed the addition of the long wave heat flux to extend on the radiative budget discussion. My primary concerns with the paper were that I felt it could've gone further in the analysis of the events (eg. how did we get to these patterns? how long-lived were they?) and attributing a more nuanced exploration of the data. Such analysis will help differentiate this study from the ample literature linking strong anticyclones/blocking to hot and dry conditions. Whether these are major or minor revisions will really be up to what the authors have at hand for their analysis.Â
Major comment: I would've liked to have seen a more clear and complete story of what lead to these events. For example, in the discussion there is a commentary on the sources of waves that could influence wave-5/6/7 patterns (eg. lines 334-338; 356-371). The authors could have also addressed this in the paper (for example through lagged composites or Hovmöller diagrams), which would have both solidified their discussion (which I felt was reaching for connection to the prior literature in places without the evidence to support it), and provided a lot more emphasis to their results. Adding to this, an exploration of the stationarity and synoptic support of these features would've been beneficial. For example, a blocking index could be applied to this work to see how well traditional blocking indices perform in identifying these features. Another example could be a lagged composite analysis of the fields in figures 5-7, which could show both what types of synoptic features helped establish these patterns (eg. was it blocking? Rossby wave breaking? diabatic reinforcement of a ridge? etc.) as well as what sort of pre-conditioning of the surface environment was in place (eg. was this a rapid-onset scenario, or was there a lead-up?).
Minor/Technical editing comments:Â
1. L40 and L42: Get rid of the " 's " after each year.Â
2. L49-51: Please define which ENSO index you're referencing here.Â
3. L73: How sensitive is your analysis to the use of 2m temperature compared to, say, 850 hPa temperature? Is the 2m T subject to biases?Â
4. L76-79: What lead to the decision to use this dataset rather than the ERA5 precipitation?Â
5. L90-91: Please provide some references to your sentence here on the SPI and it being a common index for meteorological drought.Â
6. L91-93: I found the discussion on Gringorten plotting position (and this sentence in general) unclear as written.Â
7. L105-106: Why is this a stand-alone paragraph? As written it's just a sentence.
8: L105-106: I wasn't sure on your approach here to the statistics associated with the SPI3 and the hot days index. Is your SPI3 index really Gaussian (or close to it)? Precipitation data rarely is, making your analysis approach here a bit dangerous (percentiles require a normal distribution typically).Â
9: L178-179: Is this trend driven by a single region, or by all 3? It may be interesting to look at the trends for all 3 regions. In addition, it may be helpful to look at the monthly trends/means individually (rather than how you have them shown in figure 4 (in other words, it might be helpful to decompose this by looking at it less from the annual standpoint and instead to look at the trends/statistics of each month).Â
10: L194 (and general definition of EM region): I get why you defined the regions the way you did, but it feels like an unfortunate missed opportunity that you include so much empty space (Mediterranean Sea) and don't include portions of northern Europe and the Isles. It may be worth including these?Â
11: F3: A more discrete colorbar for panel (a) would be helpful here.Â
12: L237-241: Though I do appreciate the use of upper level patterns to diagnose the flow here, I wonder about instead using a lower-tropospheric variable when linking to surface heat flux anomalies. There's a lot that can happen between 250/500 and the surface from a cloud or otherwise standpoint.Â
13: L247: You should swap the order of A3 and A4 given the order you're referencing them in the text.Â
14: F5-7: It would be helpful to add the v250 and u250 climatology, or the composites for each respective figure, to your panels g and h to supplement the discussion. Same thing for A2 (also on A2 it's really hard to see your hatching).Â
15: L264: I'm not sure what the approximately symbol is meaning in front of your greater than symbol. Please aim to be specific.Â
16: L265: Careful here (and elsewhere) to stay with scientific notation.Â
17: L269-270: Your SLHF looks less potent than over North America - does this have to do with the proximity to the Mediterranean Sea?Â
18: L333: Please flip the order of 5 and 6 for your wavenumbers hereÂ
19: L344: Is this double jet structure linked to anticyclonic Rossby wave breaking?Â
20: L380: I find the sentence starting with 'By using a fraction ...' vague and confusing - please aim to be more direct.Â
Â
Â
Citation: https://doi.org/10.5194/egusphere-2025-5073-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 312 | 241 | 38 | 591 | 25 | 28 |
- HTML: 312
- PDF: 241
- XML: 38
- Total: 591
- BibTeX: 25
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1. Clarification on Figure 5a I have a question regarding the "High FA minus Low FA" methodology mentioned in the Figure 5 caption. Since the definitions yield different sample sizes (19 vs. 92 events), does a direct subtraction of the "Number of COHDEs" create an unfair comparison? It would be helpful to know if you normalized these counts to account for the unequal sample sizes.
2. Rossby Wave Train vs. Localized Events In Figure 5g, the Rossby wave train is clearly circumglobal. I am curious why the compound events in Figure 5a appear only in North America, while the other ridges in the wave train do not seem to generate corresponding COHDEs elsewhere. Is there a specific physical mechanism dampening the response in other regions, or is this primarily due to the composite being triggered specifically by North American events?
3. Minor Correction I noticed a potential typo in Line 158. The text cites "(Fig. 1)," but based on the context, it seems it should refer to Figure 5a.